Reconceiving Military Base Redevelopment

Introduction

The U.S. Department of Defense (U.S. DOD 2010) currently possesses 4,262 military bases and defense installations in the continental United States and another 737 bases in 130 countries across the globe. The federal government spends an estimated 14% ($550 billion) of the total U.S. national budget (or $3.8 trillion) to build, support, and maintain its military base infrastructure (U.S. Office of Management and Budget [U.S. OMB] 2011). These long-term investments have altered the economic geography of the United States and other countries around the world. In many instances, the fates of cities and regions have rested, to some extent, on choices about base location, configuration, and activity along with military investments in specialized manufacturing, research and development, and technology-transfer (Kirby 1992; Markusen et al. 1991; O’Mara 2005). However, the “winners” of public and private defense investment face the prospect of losing these installations in a post–Cold War era where policy makers seek to shift defense spending away from maintaining bases and toward new methods of warfare. Over the past 30 years, the Base Realignment and Closure (BRAC) process generated five rounds of closure recommendations, in 1988, 1991, 1993, 1995, and 2005, with 17 bases closed following the first round, 28 after the second round, 32 after the third round, 32 in the fourth round, and 13 following the most recent round (U.S. DOD 2005). ¹

The primary public policy mechanism for closing bases is the BRAC program. The BRAC program endows nine commission members appointed by the U.S. president with the power to make controversial decisions about what to close or shrink based on a complex set of factors amid political pressures at the federal and local level (Freedman and Ransdell 2005; Sorenson 1998). ² The BRAC program only pertains to the most recent rounds of base closures since 1989: There have been rounds of base realignment and closures throughout U.S. history after major wars or to reconfigure force allocation to reflect new needs (e.g., the closure of long-range bomber bases in the 1960s due to the advent of intercontinental ballistic missiles [ICBMs]). The BRAC program was created in 1988 following struggles to close bases in the 1960s, 1970s, and 1980s. ³ BRAC theoretically allows for bases to close because it provides political cover for U.S. Representatives and other politicians by taking much of the closure decisions out of their hands. The DOD presents a list of bases it recommends for realignment and closure based on published criteria such as military value relative to the cost of operations as well as the impact on communities. ⁴ Next, the BRAC commissioners visit communities whose bases are listed for closure and hear testimony from potentially injured parties. The BRAC commission then has the authority to add or remove bases from its list before it makes recommendations on which bases to close. This list passes first to the president for approval and then to Congress, which has an option to disapprove the full closure list, but not to spare individual bases. Congress has never exercised this option, and the base closure list becomes law, assuming Congress fails to act at this point. The Pentagon then follows the BRAC commissions’ closure recommendations (U.S. DOD 2005). ⁵

We are interested in the BRAC closures relative to other, earlier rounds of U.S. base closures for several reasons. First, these 122 relatively recent closures have a large impact on communities in the contemporary era. Many other bases have closed in the United States in earlier eras, but such base closures are difficult to study due to lack of data and difficult to connect to current areas of community interest. Furthermore, the redevelopment challenges these earlier communities faced after their bases closed are not likely to be as relevant to communities facing base closures in 2015 and beyond. This is due to the differences in the economic, political, regulatory, and cultural context in which earlier closures occurred relative to bases that closed recently or will close in the future. Redevelopment efforts from the 1940s may indeed have lessons for communities losing a base today, but we believe the lessons from post–Cold War military redevelopment are likely to be more applicable based on relative contextual similarities on the dimensions listed above. We want to build a foundation of data to assist these communities in redeveloping their bases and therefore emphasize the most recent rounds of closures.

After the BRAC process is complete, the federal government makes deals with municipalities, quasi-public redevelopment authorities, and private developers to transfer management and/or ownership of the land and properties for eventual redevelopment and/or conservation after bases close (Business Wire 2005; Dayal 2011; Dorrier and Wiberg 1993; Gilmore 2005; Greene 1999; Lowry Community Master Association 2012; U.S. Environmental Protection Agency 2006). The sum total of these arrangements represents one of the largest transfers of capital from the federal government to local groups since urban renewal—and one of the largest transfers of land in recent U.S. history (Altshuler and Luberoff 2003; U.S. Government Accountability Office [U.S. GAO] 2007a, 2007b, 2009).

Former military bases offer communities a distinct set of redevelopment opportunities due to their existing infrastructure, such as airstrips, roads, water, electric service, and so on, stemming from their former military functions. Yet the former sites also present distinct problems when communities have to rehabilitate contaminated land stemming from previous military use (Bagaeen 2006; Davis, Hayes-Conroy, and Jones 2007; Hansen 2004; Wiley and Rhodes 1998). Moreover, communities with closed bases often have to provide services the military used to offer such as housing, utilities, or health care—in many cases with very limited resources. ⁶ Communities also have to transition these military assets/liabilities from a defense economy to a new economic foundation while mitigating the negative impact that the base closure has on local businesses. This is especially true in instances where the base was more integrated into the community rather than isolated physically and geographically (Cordero de Noriega and Gonzales 2004; Gilmore 2005).

Base closures present many similar challenges, but we currently do not have a common point of departure for addressing them. For instance, we do not know the factors driving the results of military base redevelopment projects around the country or to what extent military base redevelopment is different from any other kind of redevelopment. Current redevelopment scholarship does not incorporate the unique challenges in redeveloping former military sites. Redevelopment of military sites may suffer as a result because communities may be missing valuable knowledge about how military redevelopment works around the country. This is important because the next round of base closures is provisionally scheduled for 2015, and the Pentagon has formally requested at least one additional round after that (U.S. DOD 2013), although this may change due to recent announcement about further defense downsizing (Shanker and Cooper 2014). The rate of domestic base closures may not reach the post–Cold War peak of the early BRAC rounds, but communities where bases close will have a better chance to prepare for redevelopment if they can draw on others’ experiences—a serious challenge given the lack of data on trends for all closed bases. Moreover, many of these bases that are not undergoing conversion are struggling to make this transition and may find comparative information useful in overcoming enduring problems. This is an important argument because many communities may try to replace what they lost when bases closed without realizing the challenges they face. An initial step is gaining insight into the processes and relationships that are part of base conversion. We test arguments as to what communities that pursue master planned developments, or what we call “Replacement Community” redevelopment, have in common with one another. This information can then serve as a first step toward making policy recommendations as to the type of redevelopment a similar community facing a future base closure might pursue given the average outcomes in this area among other similar communities since their bases closed.

There are many existing studies of military base redevelopment, but few evaluations of redevelopment processes across all bases closed under the BRAC program (Bagaeen 2006; Cornell University 2013; Hill 2000; Hill, Deitrick, and Markusen 1991; Stern 2006; Urban Land Institute 2009). Extant scholarship emphasizes the immediate impact of base closures on regional economies, the problems associated with environmental remediation, questions of equity surrounding development decision making, or the supremacy of federal considerations over local decision making. Additional academic inquiries use case studies to focus on a single conceptual theme or the early stages of redevelopment processes (Bagaeen 2006; Cornell University 2013; Curtis 2011; Davis, Hayes-Conroy, and Jones 2007; Hansen 2004; Hess et al. 2001; Hill 2000; Hill, Deitrick, and Markusen 1991; Hooker and Knetter 2001; Kosla 2010; Lynch 1970). There is ample, valuable information about the redevelopment process from a few high-profile cases such as the Presidio in San Francisco or the Philadelphia Navy Yard (Cornell University 2013; Curtis 2011; Hess et al. 2001; Stern 2006; Urban Land Institute 2009). However, we do not know the extent to which these cases are representative of the challenges and opportunities facing different kinds of communities in the United States due to variation in planning cultures, market health, population and demographic forces, and geographic influences across communities. For example, a community losing a base in rural Texas may have a different set of redevelopment options at its disposal than a community losing a base in urban, coastal California.

The lack of aggregate information for all bases in the BRAC process has far-reaching implications. Communities currently plan and steer limited resources for post-BRAC redevelopment while unaware of general trends concerning military redevelopment around the country. In this article, we develop a foundation to explain some patterns of military redevelopment by explaining variation in the (re)built environment on former military bases. We gathered concrete data on all 122 shuttered bases beginning with the first BRAC round in 1989 and ending with the most recent round in 2005. The data come from the U.S. DOD, community redevelopment authorities, and publicly available documentation of redevelopment land use on city, state, and local websites. Our data set catalogs the region, geography, size, function, and branch of service of the former military base as well as the year it closed. Then we record information on an array of redevelopment inputs, information on redevelopment processes, and, finally, what appears in the physical, built environment on each former base.

The potential distinction between community choices in military base redevelopment and general redevelopment begins with partnership structures that reflect civic ambitions for physical and place transformation based on local planning cultures. For example, we collect data on the identity of the primary funding sources for redevelopment (public, private, local, state, federal, etc.), and several control variables that might influence a community’s redevelopment choices such as local market conditions or potentially high costs of remediation associated with a site’s former military function. Next, we identify and classify aspects of the built environment that appear on each site as part of its redevelopment. We divide these observations by land use into residential, commercial, industrial, institutional, and airport categories. This coding does not simply reflect zoning designations for intended future use, but what currently appears on each site. For instance, a base coded as having a residential component to its redevelopment indicates a housing development on that base—not the intention of building one in the past, the present, or the future.

We then construct a statistical model of land use in the rebuilt military base environment as a first step toward explaining the broad contours of military redevelopment. Our database allows us to identify patterns in base redevelopment stemming from the funding source for redevelopment, the number of partners in the redevelopment process, the base’s previous defense use, and the resources a community may bring to bear on redevelopment funding. At this time, we are not aware of other quantitative studies using the universe of available data to describe and explain land use in the rebuilt military environment. Our hope is that the data we collect will give policy makers and planners the benefit of information on trends covering all closed military bases, which will supplement extensive information from existing case studies. Ultimately, our research identifies broad trends in military base redevelopment and provides a supplemental direction for scholarship on the topic—one that moves beyond counting the jobs lost in the immediate aftermath of base closures—and toward a clearer understanding of military redevelopment. Identifying the determinants of redevelopment land use lays the groundwork for answering important future questions such as the extent to which communities’ redevelopment strategies are best suited to achieving their goals and stimulating long-term sustainable development. Subsequent research could then address which strategies could or should be replicated depending on general community characteristics and local conditions.

What Explains the Rebuilt Environment on Former Bases?

Recent planning research considers military issues in the context of general development, but still leaves significant gaps in our understanding of redevelopment patterns. For instance, economists and geographers explore the ways the military industrial complex has reshaped the American landscape through its direct investments in research, training, production, administrative, and support centers (Cordero de Noriega and Gonzales 2004; Markusen et al. 1991; Smith 2009). Other researchers assess the economic impact of base closures by counting jobs, services, and firms lost (Hooker and Knetter 2001; Hultquist and Petras 2012; Lynch 1970; Poppert and Herzog 2003). Historians and urban planning scholars, in turn, document the rise and fall of anchor institutions, specifically universities and medical facilities, as they chase R&D money from the DOD (Birch 2010; Hess et al. 2001; Markusen et al. 1991; O’Mara 2005). Finally, environmental scholars and energy-related policy analysts have identified the negative outcomes associated with military development such as the placement of DOD buildings and facilities in isolated areas that require the allocation of scarce natural resources (e.g., water) (Davis, Hayes-Conroy, and Jones 2007; Hansen 2004; Havlick 2011); the difficulty of cleaning up military brownfields (Davis, Hayes-Conroy, and Jones 2007; Hansen 2004) as well as concerns about environmental justice and equitable development (Davis, Hayes-Conroy, and Jones 2007).

The above research is valuable for each separate discipline and for those interested in base conversion, but it would provide a clearer picture of military base redevelopment if added to data on other aspects of the redevelopment process. For instance, studies evaluating job losses surrounding base closures in the few years after the bases close do not offer clues to resolving redevelopment puzzles—Only that jobs are gone. In contrast, our view of redevelopment includes more than jobs. We argue that an evaluation of nearly everything that happens on former bases and in former base communities over time is necessary to create a more robust framework for considering conversion resiliency. No one can evaluate everything that occurs on former bases, but studies focusing only on aggregate jobs numbers do not improve our understanding of redevelopment or resiliency as larger concepts. Moreover, the literature on military redevelopment rarely uses data from every closed base. We, therefore, do not know whether the military bases motivating these analyses are representative of the national experience. Finally, most detail-rich redevelopment literature from the BRAC round of base closures to the present does not evaluate whether general theories and assumptions of redevelopment apply in former military contexts (Hill 2000; Kosla 2010). For example, one of the more prevalent assumptions driving narratives of U.S. urban redevelopment is that the federal government is no longer a central player but an observer, an indirect funder, or an occasional participant in project-based initiatives. Analysts of modern urban revitalization pinpoint the late twentieth century as the moment when the federal government began to back out of local redevelopment activity after decades of housing and urban renewal programs that sought to modernize the city through a set planning-oriented, “command and control” formulae (Altshuler and Luberoff 2003; Fainstein 2001; Frieden and Sagalyn 1989; Sagalyn 2001). However, little is known about whether this trend necessarily holds for base conversion, where the federal government is a key stakeholder.

Early military redevelopment literature parallels general redevelopment arguments on the importance of federal involvement for achieving redevelopment goals. For instance, the U.S. DOD’s Office of Economic Adjustment (OEA) issued community redevelopment reports in the 1960s and 1970s to accompany base closures following the transition from World War II (WWII) to the Cold War. These reports were designed not only to help communities ease the process but also to aid the military and later, the federal government in distributing redevelopment grants to struggling communities (OEA 2014). However, the OEA continued to work with defense communities in the 1990s and 2000s following BRAC closures during the era of federal devolution from other types of urban redevelopment. In addition, the U.S. GAO (2007a, 2007b, 2009, 2013b) monitors and advises the DOD on how it can better assist communities in the transition. The OEA still performs helpful land-use studies and assists communities in gaining federal resources to attract jobs and generally offset resources lost when bases close, including in rural areas (Cowan 2012). Furthermore, many military redevelopment partnerships feature direct funding from the federal government and direct participation in redevelopment over the life of the redevelopment process (Houlemard 2013; Ott 2013). This raises the question of whether the federal government’s continued presence surrounding base redevelopment long after closure influences the type of projects completed on former bases.

We contend that the unique challenges of military redevelopment render the federal government’s continued presence important for project outcomes for several reasons. First, environmental rehabilitation is expensive and nearly all bases feature at least some contaminated land (Hansen 2004; U.S. GAO 2007b). The resources of the DOD and other federal agencies, such as the Environmental Protection Agency, are essential for local communities who place housing developments, commercial centers, and parks on their former bases. Second, navigating the regulatory environment to reuse land and integrate it into the community requires federal expertise and resources because many communities lack experience and resources to devote to the effort after bases close and local government revenues fall (Bennett 2011; Houlemard 2013; Ott 2013; Weiner 2013; Ziter 2013). We, therefore, hypothesize a federal presence will be associated with more complicated rebuilding efforts serving a wide variety of end users at greater rates than redevelopment where only one or two narrow uses appear on the former base.

Similarly, it is unknown whether different partnership structures or patterns characterize military redevelopment compared with general redevelopment. The federal government, state government, or local government funds some military redevelopment projects to the exclusion of any private, nongovernmental, or other public entity’s official cooperation (Patton 2013; Weiner 2013). This may be because the private market has not yet shown interest in the project, the local government lacks project resources, or the federal government retains jurisdiction over part of the former base. In contrast, other bases feature public, private, and nonprofit partnerships across different levels of government (Houlemard 2013; Ziter 2013). We believe complex redevelopment efforts likely generate complex partnerships due to the large number of stakeholders involved, the need for private capital to fund ambitious redevelopment, and the need for government partners to clear the regulatory hurdles involved in the redevelopment process. In this case, we hypothesize more extensive collaboration within redevelopment partnerships will, on average, be associated with mixed-use rebuilding at greater rates than rebuilding less complex, single-use projects. For example, a single project appearing on a site, such as an industrial center, may be possible with only private financing whereas a housing development, a commercial shopping center, a park, a hospital, and a museum are more likely to stem from collaborative efforts across public, private, and nonprofit entities.

We argue that local market conditions and the level of economic productivity in a former military community are likely to influence the variety of land uses that appear on former bases. More economically productive communities are likely to have more redevelopment options at their disposal than less productive, potentially less affluent communities due to advantages in financial resources, support from networks, leadership, and market conditions/commercial viability (e.g., the Presidio in San Francisco vs. the Alabama Ammunition Plant). These advantages may translate into systematic differences in what appears in the rebuilt environment for more productive communities with more resources relative to communities without similar advantages. For instance, private interests will be more willing to build a commercial center in more affluent rather than less affluent areas, property developers will construct houses in areas where jobs are plentiful and real estate values high, and local governments will contribute money for parks, libraries, and schools at greater rates when revenues remain high—all advantages relatively productive areas have over relatively unproductive areas. We, therefore, believe former military communities in counties with higher per capita gross domestic product (GDP) will be associated with more comprehensive, mixed-use rebuilding at greater rates than communities with lower per capita GDPs, on average. In contrast, we contend that communities pursue single-project rebuilding because relatively simple projects are low cost (e.g., converting military airstrips to commercial airports) or because the costs of pursuing the alternative of more comprehensive projects are too high in terms of environmental remediation, geographic isolation, low commercial appeal, or numerous unknown other reasons. Furthermore, we think it likely that residential, institutional, or commercial interests withdraw their proposals to participate in redevelopment once former bases are designated for airport or industrial uses because of less demand for living, recreating, or working near airports and industrial centers or for incompatible land uses. In short, communities pursue single-use projects when they exhaust other options because of high costs and increased risk in one sense or another.

We argue that the former military function of a site will influence the redevelopment costs, and, thus, the projects that ultimately appear in the rebuilt environment on a site. For instance, it is likely to be less expensive to refit an air force base for commercial or civil aviation than to demolish runways, an air tower, hangars, and so on, and replace them with an industrial manufacturing facility. Similarly, naval shipyards can become commercial shipyards with much less effort than it might take to convert the base into parks, museums, and schools. Finally, some projects might not be viable due to the environmental hazards the military leaves behind due to the function of the former base. We believe a munitions plant will be relatively difficult to convert to a residential development with schools and parks nearby due to environmental regulations and the cost of brownfield remediation. We, therefore, hypothesize that bases with high potential for environmental contamination due to their former military function (e.g., weapons plants, testing grounds, and munitions depots) will be less likely to feature mixed-use projects in the rebuilt environment than bases with less potential contamination that are therefore less costly to remediate.

We believe that time is an important factor in what appears on former bases. First, bases that closed earlier have a longer time frame to complete their projects than more recently closed bases. It is, therefore, more likely that the bases that closed earliest will feature more projects appearing on site at any given time than other bases. Second, we want to know whether the different trends seen in other areas of redevelopment, such as federal devolution of costs and responsibilities, a focus on mixed-use development, a focus on collaborative partnerships, sustainability, and so on, apply to military base redevelopment and explain the variation in land use that appears on former military bases. In particular, we are interested in change agents or leaders in charge of project development and implementation. For example, we observe informal (coalitions, alliances, collaborations that do not have a contractual agreement) and formal (i.e., contractual arrangement between different public, private, and nonprofit entities) public, private, and nonprofit partnerships to be present in 22% of all cases in the first round of closures in 1988, but 75% of cases in the most recent round of base closures in 2005. This shift corresponds with general thinking about the rise of such partnerships in a post-Reagan world of federal devolution. In contrast, the configurations of redevelopment participants most prevalent during the first round of base closures (public–private partnerships at 35%) have steadily dwindled over time to encompass just 8% of former bases in the most recent round of closures. We believe public–private partnerships to be more prevalent now than 20 years ago, but we do not know if this reflects a shift in redevelopment power away from the federal government or purely pays lip service to collaboration.

Economic development literature provides an alternative rationale for why we might expect greater redevelopment collaboration in more recent BRAC rounds. Policies creating local enterprise zones, Federal Empowerment Zones, and Enterprise Communities or state tax break legislation geared toward base conversion may target particularly troubled areas in early rounds, but relatively more productive areas later as the policy grows. The result is likely to be systematically different outcomes for early versus late participants in the program (Greenbaum 2004; Greenbaum and Bondonio 2004). Applied to BRAC rounds, it is entirely possible that early base closures were selected for their high redevelopment potential. These bases then might have attracted relatively more private capital than less attractive bases closed in later rounds and, thus, required less public or nonprofit redevelopment assistance.

We construct statistical models to test the above hypotheses and provide new insight into the determinants of one type of military redevelopment outcome—what physically appears in the rebuilt military environment. Our primary independent variables are the presence of the federal government in the redevelopment partnership; the configuration of public, private, and nonprofit entities in the partnership; and the per capita GDP of the surrounding community and include indicators for these concepts in our models. We also control for other factors likely to influence access to resources and, thus, the extent of land reuse that ultimately appears on the former base. For example, we collect data on whether the former base is near an urban environment, the geographic region where the base is located and the BRAC round when the base closed and include these variables in our statistical models. A description of our dependent variable and each independent variable follows.

Dependent Variable

Diversity of Land Use in the Rebuilt Environment

We collect data on a wide variety of land-use outcomes in military base redevelopment. We record all of the projects physically present on each former military site such as a private shipyard, a public park, a commercial shopping center, a nonprofit’s headquarters, and so on. Then we categorize these projects as one of the following land uses: Residential, Commercial, Industrial, Institutional, and Airport. At this point, we do not make distinctions between the individual projects within a former military site: We categorize parks, reservoirs, and museums as Institutional because they typically represent publicly funded, publicly used redevelopment. We prefer this approach because it provides an understanding of the general ways communities use former military land. This information also offers a valuable point of departure for evaluating subcategories or project-specific endeavors, such as parks or museums, on former military sites and highlights trends for all former military bases—not just for a few specific sites that may not represent the broader experience of military base redevelopment accurately.

We consolidate the above categories further and identify three distinct types of land use in military base redevelopment. First, approximately one-third of former military bases in the redevelopment process belong to the “Replacement Community” category, which refers to defense sites that have transformed into diverse, mixed-use communities, or large-scale master planned communities, by pursuing several kinds of projects (commercial, residential, industrial, institutional, and airport) for a wide variety of beneficiaries. The second category of land use on former military sites is “Leveraged Communities.” This category includes sites where redevelopment resulted in exactly two kinds of complementary projects. For example, land use for a site in this area includes industrial and commercial projects that promote the manufacture and subsequent sale of a product. Similarly, the category includes sites with residential and institutional land uses where redevelopment projects pair housing and parks, schools, or museums. Finally, many military sites are only redeveloped for one specific kind of land use—These “Isolated Project” sites feature exclusively residential, commercial, industrial, institutional, or airport uses. For example, sites that feature only a chemical plant or an airport would fall into this category.

Cities frequently have a goal of creating Replacement Communities, which are similar to master planned communities within an urban setting if sufficient funding, resources, and other opportunities are available. This is because Replacement Communities include a mixture of residential, commercial, institutional, and industrial uses that will serve the population, provide income for the region, and offer revenue for the city; in short, it will serve similar functions following redevelopment to those it served while operating as a military base. However, many communities understand their limitations and probably do not pursue Replacement Communities in the first place because of the low chance of completion and instead pursue more sensible options. Isolated and Leveraged Projects, respectively, contribute one and two elements to the new community, but do not “replace” it. Our categories, thus, reflect the extent to which redevelopment replaces the former base’s community in as many categories as possible—We still do not know what communities may have tried and failed to accomplish, but we can and do take stock of what presently appears on the former base.

Table 1 presents several summary statistics on military base redevelopment in the United States from a rebuilt environment perspective while the Appendix provides a map of the United States depicting base coverage by land-use type.

OPEN IN VIEWER

Table 1.

Featured Elements of U.S. Military Base Redevelopment.

Diversity of Land Use	Funding Source	Components of Projects	Physical Size of Parcels	Region	Bases in Counties with an MSA
General attributes			61 are <2 sq. miles	46 W38 S22 NE16 MW	82 border an MSA40 do not
			2 < 31 are <8 sq. miles
			30 are >8 sq. miles
Replacement Communities: 52 cases	43 feature public, private, and nonprofit partner-ships	26 have four of five land-use components26 have three of five componentsOnly 5 lack commercial land use	17 are <2 sq. miles2 < 29 are <8 sq. miles6 are >8 sq. miles	26 W16 S5 MW5 NE	46 border an MSA6 do not
Leveraged Communities: 38 cases	27 feature public–private or public–nonprofit partner-ships11 are funded by one entity	6 feature commercial–residential, 13 commercial–industrial, 6 residential–institutional land-use components13 feature other combinations	18 are <2 sq. miles2 < 19 are <8 sq. miles1 is >8 sq. miles	14 W12 S8 MW4 NE	26 border an MSA12 do not
Isolated Projects: 32 cases	28 feature public or private funding alone (single funders)	3 feature purely institutional land-use components26 feature purely industrial components5 are purely airports	17 are <2 sq. miles2 < 8 are <8 sq. miles7 are >8 sq. miles	22 W7 S2 MW1 NE	16 border an MSA16 do not

Note. MSA = metropolitan statistical area; W = West; S = South; NE = Northeast; MW = Midwest.

Independent Variables

Modeling Land Use in Military Base Redevelopment

We estimate statistical models of postmilitary land use through multinomial logit: We want to evaluate the likelihood of achieving particular categories of land-use outcomes in the rebuilt environment given the relative presence of government support; the relative collaboration between the public, private, and nonprofit sector; the former military function of the site; the community’s per capita GDP; the BRAC round when the base closed; the region where the site is located; the proximity to an MSA; and the physical size of the former base (see Table 2). The primary model uses the three broadest land-use categories as a categorical dependent variable (Replacement Communities, Leveraged Communities, and Isolated Project redevelopment). ⁸

OPEN IN VIEWER

Table 2.

Multinomial Logit Model of Diversity of Land Use in U.S. Military Base Redevelopment, 1991–2010: Isolated Project Redevelopment Is the Baseline Category.

Independent Variables	Logit Model Odds Ratio for Leveraged Projects	Logit Model Odds Ratio for Replacement Communities
Federal Funding	1.048 (0.692)	1.37** (0.466)
State Funding	0.993 (0.725)	1.15** (0.378)
Local Funding	1.03 (0.450)	1.06 (0.671)
Number of Public/Private/Nonprofit Collaborators	1.015* (0.528)	1.053** (0.340)
Environmental Hazards	0.506* (0.342)	0.198*** (0.135)
Per Capita Gross Local Product (by county, logged)	1.174** (0.329)	1.578** (0.603)
Late BRAC Round	1.023 (0.561)	1.052 (0.390)
MSA Within County	0.854 (0.425)	1.074 (0.388)
Size of the Base (square miles, logged)	1.018 (0.286)	1.036 (0.439)
Constant	−0.583 (0.840)	−0.622 (0.675)
Log likelihood	82.15
Wald χ2(3)	26.59
Number of bases	122
Pseudo-R2	.532

Note. Standard errors in parentheses are robust to heteroskedasticity. BRAC = Base Realignment and Closure; MSA = metropolitan statistical area.

*

p < .05. **p < .01. ***p < .001.

Results and Analysis

The estimates above show how federal and state redevelopment funding improve the odds of former bases featuring Replacement Community land use. We estimate that having the federal government as a funding partner increases the chances of Replacement Community outcomes by 37% ⁹ relative to Isolated Projects. This bolsters our argument that (1) federal devolution has not occurred for an important segment of communities pursuing military redevelopment and (2) a federal and state presence are critical for communities navigating complex, costly redevelopment challenges. These results are in contrast to existing literature identifying federal devolution as a prevailing trend in land redevelopment (Altshuler and Luberoff 2003; Fainstein 2001; Frieden and Sagalyn 1989; Sagalyn 2001). Next, our models show how Replacement Communities are more likely to feature multiple funding partners in the private and nonprofit sector along with multiple public entities than are Isolated Projects. We do not know the precise amount of this collaboration, but we estimate the odds of bases featuring Replacement Community land use increases by approximately 5% with each additional funding partner. This result is in keeping with our findings on state and federal funding above and is similar for multiple nonprofits as well as multiple private firms (see Table A3 in the Appendix): We argue the complex, expensive nature of replacing a former military community requires considerable expertise and funding from outside of the community, which is consistent with scholarship on the topic (Hansen 2004; Kosla 2010; U.S. GAO 2010). Similarly, it is likely communities with multiple powerful, vocal stakeholders across different sectors generate different redevelopment plans than communities with just one powerful group in the first place—particularly as more dialogue is likely to occur around denser, mixed-use projects that are near or in the vicinity of population and employment centers.

Our models demonstrate an increased likelihood of Replacement Community land-use outcomes relative to Isolated Projects as communities’ per capita GDP increases. This finding is in line with expectations about redevelopment in general: Communities with greater economic output might have stronger market conditions and more resources than less productive communities (Katz 2006; Pastor and Benner 2008). Furthermore, more productive communities may, on average, have more access to planning resources and may therefore be more inclined to follow best and coveted practices in long-term planning and trends in design by focusing on mixed-use development. Finally, isolated projects may meet the specific needs or capacities of poorer areas (e.g., a community that seeks a hospital or an airport before adding parks, housing, or even more jobs through commercial development) more often than the varied, but less imperative needs of more affluent areas. Furthermore, one Isolated Project addressing a direct community need may be easier to fund in areas with a relatively small tax base that could be further undermined by the loss of military jobs following base closures. In contrast, communities with stronger market conditions are not as likely to be so heavily dependent on taxes military families pay because of other, nonmilitary employers in the area.

Our models demonstrate a statistically significant connection between the potential presence of environmental hazards on the site of the former base and land use on that base. Bases with great potential to harbor environmental hazards, such as former weapons depots, weapons plants, and weapons testing grounds, are 50% less likely to generate Leveraged Community land use ¹⁰ and 80% less likely to feature Replacement Community land use relative to Isolated Project outcomes when compared with bases with a lower probability of environmental contamination. This is consistent with Hansen (2004) and other redevelopment research (Bagaeen 2006; Fitzgerald and Leigh 2002; U.S. GAO 2007b): Environmental hazards make residential, institutional (parks, schools, museums, etc.), and commercial redevelopment difficult to achieve due to regulatory cleanup costs and length of remediation. Instead, communities are likely to redevelop environmentally damaged sites as Isolated Projects because an industrial plant, for instance, is relatively less sensitive to the hazards nearby than other types of projects. It is, therefore, easier to target as a redevelopment goal and easier to achieve as well.

Finally, several variables are not statistically connected to land use in our data. The odds of generating different configurations of land use on former bases are not statistically different for bases that closed in early BRAC rounds compared with those that closed in the most recent, 2005 round. This is somewhat surprising considering the shift in redevelopment trends across the country over the last 20 years and the incentives provided for early military redevelopment (Greenbaum 2004; Greenbaum and Bondonio 2004). However, it may only suggest that military base redevelopment is not influenced by national redevelopment trends as much as by other factors such as market conditions, costs of environmental rehabilitation, and stakeholder power. Alternately, the lack of trends in our data could just reflect the long time it might take for broader redevelopment trends to influence military redevelopment. The lack of connection between former bases’ proximity to MSAs and their diversity of land use is also unanticipated—Scholars often conceive of rural, suburban, and urban development and redevelopment very differently (Cowan 2012; Fulton and Pendall 2001; Newburn and Berck 2006; Theobald 2001). However, pressures stemming from a community’s finances, costs of redevelopment, and influence of local stakeholders may structure redevelopment strategies across urban, suburban, and rural environments more or less equally. The same argument holds for the lack of connection between the size of a former base’s physical footprint and the diversity of land use we observe on the redeveloped base. Large bases are not systematically distinct from small bases in this area—even though large bases have more room to host more redevelopment projects. Again, this is unexpected because land area is usually part of the traditional development process for site and project selection. Other pressures aside from the base’s physical footprint motivate land-use decisions and observed outcomes in our data set.

Conclusion

This study explores the different factors associated with land use in the built environment on former military bases. The projects that physically appear on former bases only represent one type of redevelopment outcome among many. However, understanding land use on former bases is important because it sheds light on what communities actually achieve in redevelopment—not what they intended to achieve. Our data show how the presence of federal, state, and other external funding partners; the relative economic productivity of communities engaged in redevelopment; and the former function of the military base all influence land use. Our results also move the discussion surrounding general trends in military redevelopment beyond job losses and toward a clearer, comprehensive redevelopment ecology.

However, our analytic leverage is still limited: Our data do not allow us to categorize projects or processes as successful or unsuccessful because we do not have information coded on every base’s original plan. The base redevelopment program would benefit greatly from knowing the rate of project completion by comparing original project plans to what appears on the site at different times and coding the extent of any disparities between the two areas. Then one could assess whether project success stems from land-use ambitions, the type of redevelopment partnerships, or any number of other factors. This type of information may help illuminate the trade-offs between the ease of organization surrounding a concentrated interest such as a single industrial-site redevelopment bid versus the potential difficulty of organizing a bid with three or four different categories of projects. Knowing more about the relative success of redevelopment projects could then increase our understanding of concentrated versus diffuse interests in redevelopment, the relative importance of funding sources for project outcomes, and a host of other relevant areas with regard to the politics, economics, planning, and geography of military redevelopment. Our results and analysis serve as an intermediate step in this direction: We discuss what has changed over time on former bases from a physical standpoint and explain why these changes occur. Explaining the rebuilt environment, then, gives us leverage on redevelopment but comprises only one piece of the concept. We cannot speak to other aspects of redevelopment, such as the determinants of project success or completion in the aggregate yet, but our research brings us one step closer to this goal.